Yokogawa AI operates JSR chemical plant

Quick hits:

  • Yokogawa’s reinforcement learning AI technology controls the JSR chemical plant for 840 hours of operation.
  • In addition to controlling JSR plant operating parameters, AI ensured product quality while eliminating the costs associated with producing out-of-specification products.
  • Major uses of AI among manufacturers include analyzing data from controllers, manufacturing execution systems, edge and cloud applications, and drive systems.

Related to this episode:

Read the transcript below:

Welcome to Automation World’s Technology Matters site. I’m David Greenfield, Director of Content, and today I’m going to tell you how AI is now being used to autonomously run an operation at a large chemical plant and how AI is being used more widely in the industry to analyze sensor data.

So, let’s start with using AI to run the operations in the chemical plant. This test was performed by Yokogawa Electric, in conjunction with chemical manufacturer JSR Corporation, to determine the viability of AI to successfully operate a distillation column in a JSR plant. The test ran from January 17 to February 21, 2022, for a total of 840 hours, or 35 days of AI-controlled operations.

Yokogawa and JSR say the test confirmed that AI technology — in this case, a type of AI known as reinforcement learning — can control operations beyond automated PID and Advanced Process applications. Check. Yokogawa says this AI technology can address conflicting goals, such as managing the dual need to maintain high quality and achieve energy savings.

According to Yokogawa and JSR, specific achievements of the AI ​​technology during this test include: Ensuring product quality while eliminating the costs associated with producing out-of-specification products. Maintain the liquids in the distillation column at an appropriate level while maximizing the use of waste heat as a heat source. And, during bad weather that could disrupt the control status due to temperature changes, the products manufactured during these periods still met the factory’s standards.

Although the test proved that complex chemical plant operations can be successfully controlled autonomously, Yokogawa and JSR point out that there are still many situations where experienced operators must intervene. Despite this, the test really shows the development potential of AI for industrial mining applications. .

Looking at more feasible applications of AI in manufacturing, Automation World recently conducted research with end users and system integrators on their use of AI to analyze sensor data. This is a key application of the technology for any smart manufacturing or Industry 4.0 initiative, as the increasing use of sensors in these applications can quickly accumulate more data than humans will ever be able to effectively analyze. Additionally, AI is particularly good at identifying the types of anomalies that people are likely to overlook, allowing for better early detection of defects.

Our research tells us that 26% of end users are currently applying AI to sensor data for production applications. And while 26% might seem like a low number, given the relative newness of AI, this number should be considered a good early adoption rate in the application of AI in manufacturing.

When it comes to how manufacturers are using AI to analyze sensor data, systems integrators say the top three applications they see are related to controllers, manufacturing execution systems, and applications. edge and cloud. End users agreed with system integrators that controllers and MES technologies were among their top three applications of AI to sensor data, but for end users, instead of cloud technologies and peak, their other primary use was associated with drive systems.

Specifically, end users report using AI in these three areas for component and product inspection, overall asset health monitoring, identification of parts and product positions on conveyors, quality inspections and torque, temperature and vibration analyses.

Integrators also highlighted how beneficial AI can be in asset health analyses. One said: “The real benefit comes from using AI to look at all the data (vision, temperature, flow, pressure, vibration, lubrication, corrosion, etc.) to provide an assessment assessment of asset health and remaining useful life.”

Another integrator noted that if he had to pick a single application of AI technology to sensor data that would be most useful to manufacturers, it would be for vibration monitoring of rotating equipment.

So I hope you enjoyed this episode of Technology Matters. Keep watching this space for regular updates on advances and applications in industrial automation technology.